Conventionally, there has been an information output method for reading a barcode printed on a printed matter or the like, and then, outputting information such as a voice. For example, there has been a method for storing in advance information that is coincident with key information imparted to storage means, retrieving the information from a key read by means of a barcode reader, and then, outputting information or the like. In addition, there has also been proposed a technique of generating a dot pattern obtained by arranging fine dots under a predetermined rule so as to output a plenty of information or programs, picking up and digitizing the dot pattern printed on picking up the dot pattern printed onto a printed matter or the like as image data by means of a camera, digitizing, and then, outputting voice information.
However, the conventional method for outputting a voice or the like by mean of a barcode has entailed a problem that the barcode printed on a printed matter or the like is an eyesore. In addition, the above conventional method has entailed a problem that a barcode is large in size and occupies part of paper, and thus, if the barcode is thus large in size, it is impossible, on the aspect of layout, to clearly lay out a number of barcodes by partial paragraph or sentence or by character or object which has the significance appearing in images of photographs, pictures or graphics.
A dot pattern is picked up as image data by means of a camera, and then, the image data is digitized in no-color 256 gradations. In order to easily recognize dots, the variation of the gradation is differentiated, and then, a dot edge is sharpened. Next, the data on the 256 gradations is binarized to white or black. By means of this binarization, when a dot is printed on paper, a dot print error occurs, the error being caused by a print shift or blurring and a shift when a pixel has been provided. Conventionally, such a print error has been error-checked by means of a parity check. However, these error checks have had: a problem that an error check is made relative to a block of data obtained from a plurality of dots instead of a print error check by dot, making it impossible to determined with which of the dot such a print error has occurred; and a problem that a camera image picking up range must be widely determined.
Further, the above error checks have entailed a problem that a distortion occurs with the dot pattern picked up as an image exerted by a lens distortion or a distortion exerted at the time of oblique image pickup, paper expansion or contraction, medium surface curling, or printing, and then, sophisticated technique is required to correct this distortion.
The present invention has been achieved in order to solve such problems. In other words, it is an object of the present invention to provide an information input/output method using dot patterns, which is capable of imparting information or functions different depending on dots of a dot pattern displayed on a printed matter or the like, thereby defining a large amount of data by means of a dot pattern and, at the time of defining information from that dot pattern, recognizing directivity and speedily defining information; which is capable of checking an error on dot layout state; and further, which is capable of enhancing security.